Bilbao R. auteur aut IRD Ortega P. auteur aut IRD Swingedouw D. auteur aut IRD Hermanson L. auteur aut IRD Athanasiadis P. auteur aut IRD Eade R. auteur aut IRD Devilliers M. auteur aut IRD Doblas-Reyes F. auteur aut IRD Dunstone N. auteur aut IRD Ho A. C. auteur aut IRD Merryfield W. auteur aut IRD Mignot Juliette auteur aut IRD Nicolì D. auteur aut IRD Samsó M. auteur aut IRD Sospedra-Alfonso R. auteur aut IRD Wu X. auteur aut IRD Yeager S. auteur aut IRD text journalArticle eng text/pdf born digital access In recent decades, three major volcanic eruptions of different intensity have occurred (Mount Agung in 1963, El Chich & oacute;n in 1982 and Mount Pinatubo in 1991), with reported climate impacts on seasonal to decadal timescales that could have been potentially predicted with accurate and timely estimates of the associated stratospheric aerosol loads. The Decadal Climate Prediction Project component C (DCPP-C) includes a protocol to investigate the impact of volcanic aerosols on the climate experienced during the years that followed those eruptions through the use of decadal predictions. The interest of conducting this exercise with climate predictions is that, thanks to the initialisation, they start from the observed climate conditions at the time of the eruptions, which helps to disentangle the climatic changes due to the initial conditions and internal variability from the volcanic forcing.The protocol consists of repeating the retrospective predictions that are initialised just before the last three major volcanic eruptions but without the inclusion of their volcanic forcing, which are then compared with the baseline predictions to disentangle the simulated volcanic effects upon climate. We present the results from six Coupled Model Intercomparison Project Phase 6 (CMIP6) decadal prediction systems. These systems show strong agreement in predicting the well-known post-volcanic radiative effects following the three eruptions, which induce a long-lasting cooling in the ocean. Furthermore, the multi-model multi-eruption composite is consistent with previous work reporting an acceleration of the Northern Hemisphere polar vortex and the development of El Ni & ntilde;o conditions the first year after the eruption, followed by a strengthening of the Atlantic Meridional Overturning Circulation the subsequent years. Our analysis reveals that all these dynamical responses are both model- and eruption-dependent.A novel aspect of this study is that we also assess whether the volcanic forcing improves the realism of the predictions. Comparing the predicted surface temperature anomalies in the two sets of hindcasts (with and without volcanic forcing) with observations we show that, overall, including the volcanic forcing results in better predictions. The volcanic forcing is found to be particularly relevant for reproducing the observed sea surface temperature (SST) variability in the North Atlantic Ocean following the 1991 eruption of Pinatubo. specialized 066 020 15 2 501-525 2024 2190-4979 https://www.documentation.ird.fr/hor/fdi:010090600 10.5194/esd-15-501-2024 2190-4979 [F B010090600] https://www.documentation.ird.fr/hor/fdi:010090600 https://www.documentation.ird.fr/intranet/publi/2024-06/010090600.pdf Accès réservé (Intranet de l'IRD) IRD - Base Horizon / Pleins textes 2024-06-12 2025-11-06 fdi:010090600 fre